Periodic Reporting for period 1 - LuMiNouS (Next-Generation Quantum Light-Matter Interfaces based on Atom Arrays and Nanophotonic Waveguides)
Periodo di rendicontazione: 2021-04-01 al 2023-03-31
Future quantum computers could dramatically outperform computers following the classical computation paradigm on some classes of calculations, promising solutions to previously intractable problems. Likewise, quantum communication promises data transfer that is by nature impervious to eavesdropping and manipulation. Any such technology requires carriers of quantum information. Photons, individual quanta of light, are a natural choice: they are easy to route and widely used to communicate classical information. At the same time, they can also carry quantum information with low loss because they do not interact with each other.
While this is an advantage for relaying information, it is a disadvantage for information processing where photon-photon interactions are required to perform computations on the quantum information bits (qubits) the photons represent. These interactions have to be mediated via the nonlinear optical response of matter, for example individual atoms. Physical systems that have light interact with atoms in a controlled manner are called light-matter interfaces. Beyond their technological use, photon-photon interactions in light-matter interfaces can lead to nontrivial quantum manybody states of photons that are scientifically interesting in their own right. Challenges faced by any design for a light-matter interface include reliably exchanging qubits between photons and atoms, and maximizing the photon-photon interactions they mediate.
A particularly promising interface consists of individual atoms trapped in a regular array close to an optical nanofiber. Nanofibers (NF) can be fabricated from off-the-shelf glass fibers via a heating and pulling technique, without the need for complex nanofabrication. Light fields guided by the NF leak into the surrounding vacuum and interact with atoms placed there. The NF conveniently serves a double purpose: First, it helps pinning individual atoms in place by trapping them using far-detuned laser light. Second, it provides optical access to interface the atoms with near-resonant photons: photons sent through the fiber can interact with the trapped atoms, and photons subsequently re-emitted by the atoms can again be collected in the fiber.
Beyond that, NF-coupled atom arrays possess features that set them apart, including spatial order of the atoms and the promise to realize arrays with a spacing of the atoms smaller than the wavelength of resonant light. Due to the spatial order, interference effects can drastically influence photon absorption and emission. The small separation of atoms in subwavelength arrays means that the electric dipoles between neighboring atoms can directly influence each other. Atoms then do no longer act as independent scatterers, but respond as a collective to photons passing through the NF. Previous theoretical studies predicted that the combination of collective optical response and interference can result in a drastically reduced error rate for some quantum information processing tasks, and result in interesting, complex states involving multiple photons at once.
Any experimental demonstration of these effects has to come to term with real-world imperfections: at present, not every trap site in a NF-based atom array is actually filled with an atom during the loading procedure, and atoms are not perfectly pinned but can move in their traps. These imperfections are likely to negatively impact both the interference effects due to order, and the collective optical response due to a close spacing of the atoms. The aim of this project was to investigate the impact of these imperfections and to mitigate and potentially exploit them as a resource.
While this is an advantage for relaying information, it is a disadvantage for information processing where photon-photon interactions are required to perform computations on the quantum information bits (qubits) the photons represent. These interactions have to be mediated via the nonlinear optical response of matter, for example individual atoms. Physical systems that have light interact with atoms in a controlled manner are called light-matter interfaces. Beyond their technological use, photon-photon interactions in light-matter interfaces can lead to nontrivial quantum manybody states of photons that are scientifically interesting in their own right. Challenges faced by any design for a light-matter interface include reliably exchanging qubits between photons and atoms, and maximizing the photon-photon interactions they mediate.
A particularly promising interface consists of individual atoms trapped in a regular array close to an optical nanofiber. Nanofibers (NF) can be fabricated from off-the-shelf glass fibers via a heating and pulling technique, without the need for complex nanofabrication. Light fields guided by the NF leak into the surrounding vacuum and interact with atoms placed there. The NF conveniently serves a double purpose: First, it helps pinning individual atoms in place by trapping them using far-detuned laser light. Second, it provides optical access to interface the atoms with near-resonant photons: photons sent through the fiber can interact with the trapped atoms, and photons subsequently re-emitted by the atoms can again be collected in the fiber.
Beyond that, NF-coupled atom arrays possess features that set them apart, including spatial order of the atoms and the promise to realize arrays with a spacing of the atoms smaller than the wavelength of resonant light. Due to the spatial order, interference effects can drastically influence photon absorption and emission. The small separation of atoms in subwavelength arrays means that the electric dipoles between neighboring atoms can directly influence each other. Atoms then do no longer act as independent scatterers, but respond as a collective to photons passing through the NF. Previous theoretical studies predicted that the combination of collective optical response and interference can result in a drastically reduced error rate for some quantum information processing tasks, and result in interesting, complex states involving multiple photons at once.
Any experimental demonstration of these effects has to come to term with real-world imperfections: at present, not every trap site in a NF-based atom array is actually filled with an atom during the loading procedure, and atoms are not perfectly pinned but can move in their traps. These imperfections are likely to negatively impact both the interference effects due to order, and the collective optical response due to a close spacing of the atoms. The aim of this project was to investigate the impact of these imperfections and to mitigate and potentially exploit them as a resource.
In the course of the project, theoretical investigations into nanofiber-coupled subwavelength arrays of atoms were carried out, using analytical and numerical tools. There were two principal results:
1. We identified tell-tale signatures of the onset of a collective optical response in the reflectance spectra of imperfectly filled atom arrays. Already short chains containing on the order of 10 atoms with subwavelength separation behave drastically different than 10 independent scatterers. If by chance such a chain is created close to the beginning of a trap array during the atom loading procedure, its presence is revealed by resonances in the reflectance spectrum that are absent if there is no collective response. Such a reflectance spectrum can be obtained by sending a weak probe laser with tunable frequency through the nanofiber, and recording how much light is reflected back as a function of the driving frequency.
2. We identified a potential method for purifying arrays with holes to obtain defect-free chains despite random atom loading. Individual atoms or collections of few atoms in neighboring sites scatter light most efficiently when it is resonant with the ground-to-excited-state transition of the atom species. In each scattering event, these atoms have the chance to recoil due to the momentum carried by the photon they interact with, an effect called recoil heating. Over time, they can pick up sufficient kinetic energy to overcome their trap potential and leave the array. Subwavelength chains of atoms, on the other hand, cannot interact efficiently with light at the same frequency due to their collective behavior. This suggest that these chains are protected from recoil heating, and irradiating a defective array with resonant light removes unwanted isolated atoms and the shortest chains of consecutive atoms first.
1. We identified tell-tale signatures of the onset of a collective optical response in the reflectance spectra of imperfectly filled atom arrays. Already short chains containing on the order of 10 atoms with subwavelength separation behave drastically different than 10 independent scatterers. If by chance such a chain is created close to the beginning of a trap array during the atom loading procedure, its presence is revealed by resonances in the reflectance spectrum that are absent if there is no collective response. Such a reflectance spectrum can be obtained by sending a weak probe laser with tunable frequency through the nanofiber, and recording how much light is reflected back as a function of the driving frequency.
2. We identified a potential method for purifying arrays with holes to obtain defect-free chains despite random atom loading. Individual atoms or collections of few atoms in neighboring sites scatter light most efficiently when it is resonant with the ground-to-excited-state transition of the atom species. In each scattering event, these atoms have the chance to recoil due to the momentum carried by the photon they interact with, an effect called recoil heating. Over time, they can pick up sufficient kinetic energy to overcome their trap potential and leave the array. Subwavelength chains of atoms, on the other hand, cannot interact efficiently with light at the same frequency due to their collective behavior. This suggest that these chains are protected from recoil heating, and irradiating a defective array with resonant light removes unwanted isolated atoms and the shortest chains of consecutive atoms first.
The combination of spatial order and collective optical response in a subwavelength atomic array is a powerful resource for quantum information processing. However, experimental demonstrations of the collective response are challenging, since completely and reliably filling all trap sites in a nanofiber-based atom array is currently technologically not feasible. But missing atoms in an imperfectly prepared array of trapped atoms interrupt the spatial order and prevent a collective optical response of all atoms present. Results of this project indicate that tell-tale resonances in the reflectance spectrum of an imperfect array randomly populated with atoms can serve as a simple way of experimentally proving the presence of collective effects, without the need of first preparing large, defect-free arrays. This is possible because even comparatively short arrays of around 10 atoms already exhibit the characteristics of spatial order and collectivity.
In order to exploit these effects for quantum information processing applications, it is however not sufficient to detect their influence even in defective arrays: it is necessary to isolate the corresponding short defect-free arrays by removing the other, scattered atoms also trapped on the nanofiber. The suggested purification technique via resonant recoil heating could allow to distill defect-free atom arrays of sufficient size from a larger, randomly assembled and defective ensemble of trapped atoms, removing the difficult requirement of perfect filling of all trap sites during the atom loading procedure. If implemented successfully, this technique could thus accelerate the exploitation of the unique properties of nanofiber-coupled subwavelength atom arrays to explore novel manybody states of light and approaches to quantum information processing with unprecedented fidelity.
In order to exploit these effects for quantum information processing applications, it is however not sufficient to detect their influence even in defective arrays: it is necessary to isolate the corresponding short defect-free arrays by removing the other, scattered atoms also trapped on the nanofiber. The suggested purification technique via resonant recoil heating could allow to distill defect-free atom arrays of sufficient size from a larger, randomly assembled and defective ensemble of trapped atoms, removing the difficult requirement of perfect filling of all trap sites during the atom loading procedure. If implemented successfully, this technique could thus accelerate the exploitation of the unique properties of nanofiber-coupled subwavelength atom arrays to explore novel manybody states of light and approaches to quantum information processing with unprecedented fidelity.